Does higher Institutional Quality improve the Appropriateness of Healthcare Provision?
Introduction
Institutions matter. From the seminal work of North, 1981, North, 1990, North, 1991, the role of institutions in affecting behaviors in society has been widely recognized. Institutions influence economic performance through various mechanisms.1 The institutional framework affects investment and growth (Mauro, 1995), the effectiveness of foreign aid and the expected benefits of natural resource endowments (Burnside and Dollar, 2000, Mehlum et al., 2006), firms’ efficiency (Dal Bó and Rossi, 2007), and even the emergence of organized crime (Acemoglu et al., 2020).
In the public sector, where public officials need to be kept accountable with implicit incentives (e.g., Alesina and Tabellini, 2007, Alesina and Tabellini, 2008), the role of the institutional framework is of primary importance. Mauro (1998) studies the composition of public expenditure across over 100 countries, and finds that countries where the perception of corruption is higher spend less on sectors which provide less lucrative opportunities for public officials. Similarly, Gupta et al. (2001) report that corruption is associated with higher military spending, suggesting the presence of rent-seeking behavior. In the political sector, Nannicini et al. (2013) show that the electoral punishment of politicians’ misbehaviors (such as absenteeism in Italian Parliament) is considerably larger in districts with higher social capital. Wong et al. (2017) examine the effects of an institutional reform implemented in China, and find that a higher quality of governance in rural villages increases the quality of public infrastructure provision.
In this study we investigate the effect of institutional quality on the appropriateness of healthcare provision in Italian hospitals. Healthcare spending in OECD countries represents about 9% of GDP and the majority of health expenditure is publicly financed (OECD, 2017). Healthcare accounts for about 15% of total public spending, though in some countries such as the United States and Germany more than 20% of public spending is for healthcare (OECD, 2017). Health spending is expected to grow further driven by an ageing population and technological innovation, raising challenges to the sustainability of public finances. Increasing the appropriateness in healthcare provision is thus critical in making health systems sustainable.
Due to the asymmetry of information which characterizes the doctor-patient relationship (Arrow, 1963), healthcare is vulnerable to providers’ opportunistic behaviors, which may lead to under- or over-treatment (Ellis and McGuire, 1986). This is also the case within the hospital sector, where providers respond to financial incentives across a range of health systems, including those based on a National Health Service where hospital physicians are typically salaried (see Section 3 for more detailed discussion). The asymmetry of information is further exacerbated by the difficulty for public insurers to measure providers’ performance, which further reduces their accountability (European Commission, 2017).
We focus on cesarean section rates during childbirth in Italy as a measure of appropriateness of healthcare provision. Cesarean section rates for first-time mothers are recognized as a valid indicator of appropriateness in the provision of healthcare services both in the literature and by policymakers (Baicker et al., 2006, OECD, 2009). In absence of clinical reasons or complications (e.g., uterine rupture), vaginal delivery is the recommended mode of delivery. Cesarean section is more invasive, it involves risks during surgery, has longer recovery times following birth and is more costly. For example, cesarean delivery has been found to increase the risk of maternal mortality and re-hospitalization, infant morbidity, and is negatively associated with child cognitive development (Lydon-Rochelle et al., 2000, Villar et al., 2006, Li et al., 2014, Polidano et al., 2017).
Crucially, the decision making in childbirth falls in the gray area of medicine, implying that it is possible for physicians to argue that a cesarean section is appropriate when it could be avoided (Chandra et al., 2011, Johnson and Rehavi, 2016). This makes the choice of childbirth delivery particularly exposed to the discretion of physicians (Gruber and Owings, 1996, Hopkins, 2000, Di Giacomo et al., 2017) and to what in the health economics literature is known more broadly as “supplier-induced demand” (Rice, 1983).2
There are several mechanisms through which local institutional quality can affect the appropriateness of hospital services. Regional authorities can measure and monitor caesarean sections for each hospital, publish them in the public domain to “shame” providers with excessive rates or to trigger audits and ask for detailed accounts of their treatment choices. They can also reduce the price differential between a cesarean section and a natural delivery, to dampen or eliminate the financial incentives. The extent to which these measures are introduced will depend on the quality of local administration, their accounting systems, and their determination and culture to reduce waste of public resources. Aside from formal measures, a local institutional environment characterized by widespread corruption, poor governance, and weak rule of law can reinforce hospital incentives to over-treat.
Italy exhibits large variations in cesarean section rates across regions that cannot be explained by sociodemographic and clinical factors (e.g., Francese et al., 2014, Guccio and Lisi, 2016). In turn, this has led to concerns by the Italian Ministry of Health that regularly monitors cesarean section rates across regions and providers (Ministry of Health, 2017). The cesarean section rate in Italy was 35% in 2016 and varied between 20% in Trentino Alto Adige and 59% in Campania, well beyond the level deemed appropriate by the World Health Organization (1985) of around 20%, as for instance observed in Nordic countries (OECD, 2017).
We measure institutional quality in Italian regions with the Institutional Quality Index, a multidimensional quantitative indicator (Nifo and Vecchione, 2014) based on the hierarchy framework employed by the World Bank Worldwide Governance Indicators (Kaufmann et al., 2010). As an alternative measure, we also employ a perception-based multidimensional indicator, the European Quality of Government Index (EQI) (Charron et al., 2014).3 Indicators of the quality of institutional environment (e.g., corruption, quality of government) display large differences across Italian regions (Golden and Picci, 2005, Charron et al., 2014). For example, Charron et al. (2014, p. 74) notice that “the gap between Bolzano and Campania in the data is much larger than the gap in the national averages between Denmark and Portugal”. Such regional variation makes Italy a particularly suitable case study to investigate the effect of institutional quality (e.g., Nannicini et al., 2013, Castro et al., 2014, Lasagni et al., 2015).
To identify the causal effect of institutional quality on the appropriateness of healthcare services, we employ an instrumental variable approach based on historical data. We follow the influential literature which argues that current institutional backwardness is also a byproduct of history (e.g., North, 1981, Acemoglu et al., 2001, Tabellini, 2010, Guiso et al., 2016): current differences in institutional quality across Italian regions have been shaped by different cultural and political histories. Specifically, our instruments are an indicator of quality of political institutions in the period from 1600 to 1850 and the percentage of population over age 6 able to read in 1881, provided in Tabellini (2010).
Our results show that institutional quality improves the appropriateness in the provision of childbirth services. A standard deviation increase in our indicator of institutional quality leads to a reduction of about 10 percentage points in cesarean section rates. The results are robust to alternative indicators of institutional quality and samples. We also find that, amongst the different dimensions of institutional quality, corruption and government effectiveness in the local area appear to be the most relevant in affecting the provision of childbirth services.
We provide evidence on possible mechanisms through which institutional quality affects cesarean section rates. Among them, we find that higher institutional quality maps into pro-active policies (i.e. hospitals’ payment policies) to reduce inappropriate provision, which in turn lead to lower cesarean section rates. We also show that institutional quality does not lead to worse mothers’ health outcomes as measured by emergency readmissions. Finally, we investigate if institutional quality affects not only appropriateness of care, as measured by cesarean section rates, but also clinical quality. We show that higher institutional quality reduces heart attack mortality rates, but does not affect stroke mortality or mortality for chronic obstructive pulmonary disease.4 Overall, our results suggest that a worse institutional environment weakens hospitals’ incentives to select treatments that are cost-effective.
Our study brings together two strands of the literature. First, we contribute to the literature on the effect of the quality of institutional environment in the public sector (e.g., Mauro, 1998, Nannicini et al., 2013, Castro et al., 2014, Wong et al., 2017) by highlighting the importance of the health sector. Second, we contribute to the literature on the causes of inappropriateness in healthcare provision (e.g., Brown, 1996; Dubay et al., 1999, Gruber et al., 1999, Currie and MacLeod, 2008, Epstein and Nicholson, 2009, Francese et al., 2014, Johnson and Rehavi, 2016, Foo et al., 2017) by highlighting the role of institutional quality in affecting physician behavior.
Only very few studies have investigated the role of institutional quality in the healthcare sector. Di Tella and Schargrodsky (2003) analyze the prices paid by public hospitals for basic inputs during a crackdown on corruption in Buenos Aires, finding that prices decreased by about 15% relative to the pre-crackdown period. Azfar and Gurgur (2008) study healthcare provision in the Philippines, and report that a higher corruption in municipalities decreases immunization rates, delays vaccination of newborns and increases waiting time. Cavalieri et al. (2017) study the execution of public contracts for infrastructure in the hospital sector, and show that the performance in the provision of infrastructure is negatively affected by environmental corruption. With respect to this literature, we focus on inappropriate behavior in healthcare provision rather than prices or the execution of infrastructures in the hospital sector. Moreover, we also show that, while corruption is among the most relevant environmental factors, it is not the only source of waste in healthcare.
The study, which is closest to ours, is Francese et al. (2014). It investigates the impact of institutional features, such as supply and pricing policies, and some political economy indicators (namely, occupation and years of experience of the regional president), on regional cesarean section rates in Italy over the period 1998–2005. In their regional level analysis, they include regional fixed effects and thus exploit variation over time to identify the determinants of cesarean sections. In our study, we investigate a range of dimensions of local institutional quality in which hospitals operate. As shown below, such institutional quality varies little over time, while it varies significantly across regions.5 We therefore exploit variation across regions and employ an instrumental variable strategy to address the potential omitted-variable bias. Our unit of analysis is at the more disaggregated hospital level rather than regional level (though institutional quality varies across regions), and we use the more recent data for the period 2007–2012.
The rest of the study is organized as follows. Section 2 describes the Italian healthcare system. Section 3 provides a conceptual framework which discusses the mechanisms through which local institutional quality may affect the appropriateness of healthcare provision. Section 4 describes the data. Section 5 provides the empirical strategy. Section 6 gives the results. Section 7 concludes and draws implications for healthcare policy.
Section snippets
The Italian healthcare system
The Italian National Health Service (NHS) was established in 1978 as a public health system providing universal access for a large basket of healthcare services (known as Livelli Essenziali di Assistenza). During the 1990s, a major reform was undertaken which separated the purchasers (local health authorities) and the providers of health services, a form of quasi-market model, and gave providers more autonomy in managing their costs (France et al., 2005). Following the reform, patients have
Conceptual framework
An extensive theoretical literature has modeled provider behavior in the healthcare sector. The seminal paper by Ellis and McGuire (1986) shows how healthcare providers, such as hospitals, choose the intensity of care to trade-off altruistic concerns against profit. Hospitals are complex organizations. The payoff function is generally modeled with a reduced-form approach that encompasses the utility functions of key workers, such as doctors, managers (Chalkley and Malcomson, 1998a, Chalkley and
Hospital level data
We merge data from different sources. The primary source is the National Program for Outcome Assessment (Programma Nazionale Esiti, PNE), which is funded by the Italian Ministry of Health and the National Agency for Regional Health Services (AGENAS). The PNE was developed to provide reliable comparative assessment of healthcare providers with respect to the effectiveness, equity and appropriateness of treatments.7
Empirical specification
Our empirical strategy aims at estimating the causal effect of institutional quality on cesarean section rates. Our regression model iswhere is the risk-adjusted cesarean rate in hospital i in region r at time t, is the institutional quality in region r at time t, is a vector of control variables at the hospital level (number of beds, birth deliveries, hospital type, whether the hospital is located in the
Baseline estimates
Table 4 provides our baseline results. For each estimate, we report bootstrapped standard errors clustered at regional level. Column (1) in Table 4 provides the OLS model. It suggests that higher institutional quality, as measured by IQI, reduces cesarean section rates. The effect is statistically significant at 1% level. The coefficient of −0.35 implies that a standard deviation increase in our indicator of institutional quality (equal to 0.21) reduces cesarean section rates by about 7
Conclusions
We have investigated the effect of institutional quality on the appropriateness of healthcare provision, as measured by the excess use of cesarean sections in childbirth delivery. We found that higher institutional quality does improve the appropriateness in the provision of childbirth services in Italy, and the effect is both statistically and economically significant.26
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (119)
Reward structures and the allocation of talent
Eur. Econ. Rev.
(1995)- et al.
Bureaucrats or politicians? Part II: Multiple policy tasks
J. Public Econ.
(2008) - et al.
To bid or not to bid: That is the question: Public procurement, project complexity and corruption
Eur. J. Polit. Econ.
(2016) - et al.
Quality competition with profit constraints
J. Econ. Behav. Org.
(2012) Physician demand for leisure: implications for cesarean section rates
Journal of Health Economics
(1996)- et al.
On the role of environmental corruption in healthcare infrastructures: An empirical assessment for Italy using DEA with truncated regression approach
Health Policy
(2017) - et al.
Contracting for health services when patient demand does not reflect quality
J. Health Econ.
(1998) - et al.
Who ordered that? The economics of treatment choices in medical care
- et al.
Corruption and inefficiency: Theory and evidence from electric utilities
J. Public Econ.
(2007) - et al.
The determinants of corruption in Italy: Regional panel data analysis
Eur. J. Polit. Econ.
(2007)
The impact of malpractice fears on cesarean section rates
J. Health Econ.
Provider behavior under prospective reimbursement: Cost sharing and supply
Journal of Health Economics
The formation and evolution of physician treatment styles: an application to cesarean sections
J. Health Econ.
Tropics, germs, and crops: how endowments influence economic development
J. Monet. Econ.
Understanding inappropriateness in health spending: The role of regional policies and institutions in caesarean deliveries
Reg. Sci. Urban Econ.
The effects of hospitals’ governance on optimal contracts: Bargaining vs. contracting
J. Health Econ.
Physician fees and procedure intensity: the case of cesarean delivery
J. Health Econ.
Thus do all. Social interactions in inappropriate behavior for childbirth services in a highly decentralized healthcare system
Reg. Sci. Urban Econ.
Corruption and military spending
Eur. J. Polit. Econ.
Are Brazilian women really choosing to deliver by cesarean?
Soc. Sci. Med.
Manipulation and auditing of public sector contracts
Eur. J. Polit. Econ.
The effects of mode of delivery and time since birth on chronic pelvic pain and health-related quality of life
Int. J. Gynecol. Obstetr.
Corruption and the composition of government expenditure
J. Public Econ.
Heterogeneous effects of patient choice and hospital competition on mortality
Soc. Sci. Med.
Do financial incentives trump clinical guidance? Hip Replacement in England and Scotland
J. Health Econ.
Weak states: Causes and consequences of the Sicilian Mafia
Rev. Econ. Stud.
The colonial origins of comparative development: An empirical investigation
Am. Econ. Rev.
The rise of Europe: Atlantic trade, institutional change, and economic growth
Am. Econ. Rev.
Democracy does cause growth
J. Polit. Econ.
Persistence of power, elites, and institutions
Am. Econ. Rev.
Bureaucrats or politicians? Part I: a single policy task
Am. Econ. Rev.
The Civic Culture: Political Attitudes and Democracy in Five Nations
Uncertainty and the welfare economics of medical care
Am. Econ. Rev.
Does corruption affect health outcomes in the Philippines?
Econ. Govern.
Geographic variation in the appropriate use of cesarean delivery
Health Aff.
The Moral Basis of a Backward Society
Decentralization of governance and development
J. Econ. Perspect.
Economic growth in a cross section of countries
Q. J. Econ.
Hospital volume and surgical mortality in the United States
N. Engl. J. Med.
The impact of competition on management quality: evidence from public hospitals
Rev. Econ. Stud.
The origins and political consequences of social capital
Br. J. Polit. Sci.
State history and economic development: evidence from six millennia
J. Econ. Growth
The empirics of growth: An update
Brook. Pap. Econ. Act.
Symmetric and asymmetric effects of proximities. The case of M&A deals in Italy
J. Econ. Geogr.
Individual interactions, group conflicts, and the evolution of preferences
Aid, policies, and growth
Am. Econ. Rev.
Bootstrap-based improvements for inference with clustered errors
Rev. Econ. Stat.
An assessment of the waste effects of corruption on infrastructure provision
Int. Tax Public Fin.
Financial incentives and inappropriateness in health care: Evidence from Italian Cesarean Sections
FinanzArchiv: Public Fin. Anal.
Contracting for health services with unmonitored quality
Econ. J.
Cited by (20)
Regional institutional quality and territorial equity in LTC provision
2023, Journal of the Economics of AgeingInternational ambidexterity and innovation performance: The moderating role of the host country's institutional quality
2022, Journal of Innovation and KnowledgeCitation Excerpt :However, some studies have used different measurements. De Luca et al. (2021) used the institutional quality index proposed by Nifo and Vecchione O to measure institutional quality at the regional level. This study followed the mainstream view and drew on the scale proposed by Busenitz et al. (2000), which is widely used to measure the quality of the institutional environment and has good reliability and validity.
Hospital quality interdependence in a competitive institutional environment: Evidence from Italy
2021, Regional Science and Urban EconomicsFood insecurity, environment, institutional quality, and health outcomes: evidence from South Asia
2024, Globalization and Health